Article Per Year (5 Year)
p-Index From 2021 - 2026
3.03
P-Index
This Author published in this journals
All Journal Jurnal Inotera Jurnal Informatika Universitas Pamulang JURIKOM (Jurnal Riset Komputer) Jurnal Administrasi dan Manajemen JISKa (Jurnal Informatika Sunan Kalijaga) JAMAIKA: Jurnal Abdi Masyarakat Abdi Jurnal Publikasi Journal of Artificial Intelligence and Digital Business LOGIC : Jurnal Ilmu Komputer dan Pendidikan Biner : Jurnal Ilmu Komputer, Teknik dan Multimedia OKTAL : Jurnal Ilmu Komputer dan Sains BULLET : Jurnal Multidisiplin Ilmu APPA : Jurnal Pengabdian Kepada Masyarakat Jurnal Ilmu Komputer dan Sistem Informasi Journal of Innovative and Creativity JUPIKMAS Jurihum
Ahmad Fauzi
Universitas Pamulang
Humanities Astronomy Chemistry Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Education Electrical & Electronics Engineering Engineering Industrial & Manufacturing Engineering Languange, Linguistic, Communication & Media Law, Crime, Criminology & Criminal Justice Library & Information Science Materials Science & Nanotechnology Mathematics Mechanical Engineering Medicine & Pharmacology Nursing Social Sciences Transportation Other

Published : 22 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 22 Documents
Search

 1   2   3 
Optimasi Hyperparameter XGBoost Berbasis Bayesian Optimization dan SHAP untuk Prediksi Dropout Mahasiswa Perguruan Tinggi Herwis Gultom; Ahmad Fauzi; Ria Ester
JURIKOM (Jurnal Riset Komputer) Vol. 13 No. 4 (2026): Agustus 2026
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/jurikom.v13i4.10055

Abstract

Putus studi (dropout) mahasiswa merupakan masalah pendidikan tinggi yang merugikan secara finansial dan sosial, terutama karena proporsi kelas dropout yang tidak seimbang serta minimnya model prediksi yang transparan dan teruji stabil. Penelitian ini menerapkan algoritma Extreme Gradient Boosting (XGBoost) yang hyperparameter-nya dioptimasi melalui Bayesian Optimization berbasis Gaussian Process sebanyak 50 iterasi, dilengkapi ambang keputusan berbasis skor F2, yaitu metrik yang memberi bobot lebih besar pada recall dibandingkan precision, untuk mengatasi ketidakseimbangan kelas, serta interpretasi Shapley Additive exPlanations (SHAP) guna menjelaskan kontribusi tiap fitur terhadap prediksi model. Penelitian ini bertujuan mengembangkan kerangka kerja prediksi dropout yang akurat, transparan, dan tervalidasi secara statistik pada dataset 4.424 mahasiswa dengan 34 atribut akademik, demografis, dan sosioekonomi. Hasil sementara menunjukkan bahwa model XGBoost teroptimasi mencapai luas area di bawah kurva Receiver Operating Characteristic (ROC-AUC) sebesar 0,930 dan luas area di bawah kurva presisi-recall (PR-AUC) sebesar 0,904 pada data uji, dengan recall 90,49% pada ambang F2 optimal sebesar 0,2251. Validasi tambahan melalui nested cross-validation menghasilkan ROC-AUC rata-rata 0,924, yang mengonfirmasi stabilitas performa model. Interpretasi SHAP mengungkapkan bahwa jumlah mata kuliah yang lulus pada semester kedua merupakan prediktor paling dominan terhadap risiko dropout. Temuan ini menegaskan bahwa integrasi optimasi hyperparameter dan explainable AI mampu menghasilkan model prediksi dropout yang andal dan akuntabel bagi kebijakan intervensi akademik dini.
Benchmarking Deep Learning Models for Multi-Class Fruit Quality and Formalin Contamination Detection in Horticultural Fruits Ahmad Fauzi; Achmad Lutfi Fuadi; Agus Heri Yunial
Jurnal Inotera Vol. 11 No. 2 (2026): July - December 2026
Publisher : LPPM Politeknik Aceh Selatan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31572/inotera.Vol11.Iss2.2026.ID732

Abstract

The adulteration of fresh horticultural produce with formaldehyde (formalin) and the rapid physiological decay of fruit remain persistent food-safety and economic problems within agricultural supply chains. Deep learning offers a non-destructive route to automated fruit-quality screening; however, the literature has largely pursued peak accuracy using progressively heavier network backbones, leaving the computational cost of that accuracy and therefore the feasibility of on-site, embedded deployment insufficiently examined. This study benchmarks three representative visual-recognition paradigms of markedly different scale, namely MobileNetV2 (a depthwise-separable lightweight convolutional neural network), ResNet-50 (a deep residual network), and ViT-B/16 (a patch-based Vision Transformer), on the identical three-class task of distinguishing fresh, rotten, and formalin-mixed fruit. A total of 10,154 images from the public FruitVision dataset were processed under a single fixed preprocessing and training protocol, so that architecture was the only experimental variable. Rather than accuracy alone, performance was assessed through an efficiency lens that normalises accuracy by parameter count and by inference cost (GFLOPs). The lightweight MobileNetV2 attained the highest test accuracy (99.69%) while using approximately one-seventh of the parameters and one-thirteenth of the floating-point operations of ResNet-50, and about one twenty-fifth of the parameters of ViT-B/16; the Transformer recorded the lowest accuracy (99.28%) at the highest cost. These results indicate that, for this medium-scale task, additional architectural capacity yields a negligible accuracy return, and that a compact model can support reliable, safety-critical formalin screening on resource-constrained devices. Per-class reliability and deployment implications, including one residual false negative, are reported without overstatement.
Co-Authors Abu Khalid Rivai Achmad Casdik Achmad Lutfi Fuadi Ade Napila Agus Heri Yunial Ahzril Pria Adistya Akbar Valentino Aldie Hendrawanto Andrey Gustaph Setiawan Andrian Hidayat Bangun Suprianto Chandra Alim Dede Eko Saputro Defri Salsabila Oktaviani Desi Sihamita Dhamar Galih Alfarizi Diski Apriana Fathih Sulthoni Nabhan Herwis Gultom Irfan Alamsah Iwan Giri Waluyo Jepri Afrizal Khoiriyah Lisda Pebriyanti Luki Handoko Mardiyanto Mochamad Aminnur Muhammad Fakri Wahidin Muhammad Farras Ma’ruf Perani Rosyani Radja Susun Pakpahan Raihan Akbar Syamputra Reza Saputra Ria Ester Risa Nur Aisiyah Safira Madinah Widia Safitri, Andin Eka Sofyan Mufti Prasetiyo Sutisna Mustika Rahmat Thoty Ariesandes Tukiyat Tukiyat Wike Rahayu Yoga Febi Atmaja Yoggi Ilham Mulya Yusup Ramadan Zein Albi Gunawan